L I B RAR.Y
57 2. .05
FIELDIANA . ANTHROPOLOGY
CHICAGO NATURAL HISTORY MUSEUM
Volume 36 October 20, 1961 No. 10
A DEFINITION OF PRIMITIVE ART 1
Phillip H. Lewis
Curator, Primitive Art
Increasing popular and academic interest in primitive art is pro-
ducing a need for accurate delineation of the term. When one sees
the various exhibitions of arid publications on primitive art it is
apparent that there is much confusion over its meaning and limi-
tations. It is the aim of this paper to define primitive art in ways
meaningful to and operational for both anthropology and history
At Chicago Natural History Museum a Hall of Primitive Art has
been opened. On a practical level, it is necessary to select specimens
for the new hall from the approximately one-half million material
artifacts in the Museum's archaeological and ethnological collections.
To do this one must first be able to differentiate between art and
non-art, and second, to separate primitive art from other kinds of art.
Primitive art is comprised of art objects made and used by mem-
bers of primitive societies. First, let us define the terms "art" and
ART AND NON-ART
The utility, or function, of art is part of the European concept of
art and leads to ethnocentric bias when Europeans consider non-
European art. The use to which the maker of an art object intends
it to be put delineates the dividing line between so-called "fine"
(pure, serious) art and "minor" (applied, decorative, industrial,
commercial, etc.) art.
Even a superficial view of European art history suggests that art
objects considered to be fine art today, as, for instance, Praxiteles'
sculptured marble figure, Hermes with the Infant Dionysus (Gard-
1 This paper was read in part at the annual meeting of the American Anthro-
pological Association, in November, 1960.
Library of Congress Catalog Card Number: 61-17857
No. 931 221
222 FIELDIANA: ANTHROPOLOGY, VOLUME 36
ner, 1936, fig. 253, p. 170), were made and used as articles of religious
paraphernalia. Since medieval times there has been tremendous
secularization of European art; the so-called fine arts now have a
secular function instead of a religious one. At present most church
art is uninspired, almost factory-made work.
It is clear that most, if not all, objects of primitive art suffer
diminution of art value if the criterion of creation for art's sake alone
be invoked. The following quotations from two writers suggest that
the phrase "art for art's sake" implies that art transcends and modi-
fies the utility of the object itself.
Erwin Panofsky (1955, p. 11) says: "It is possible to experience
every object, natural or man-made, aesthetically. We do this . . .
when we just look at it . . . without relating it, intellectually or emo-
tionally, to anything outside of itself."
Also, he says (loc. cit.) that although natural objects may be ex-
perienced aesthetically, a "man-made object . . . either demands or
does not demand to be so experienced, for it has ... an 'intention.' "
Panofsky divides all man-made objects into two classes: (1) works
of art, which demand to be experienced aesthetically, and (2) "prac-
tical" objects, which do not so demand. Practical artifacts may be
divided into: (a) vehicles of communication and (b) tools or appa-
ratuses. All man-made objects have "intent" but only art objects
demand to be experienced aesthetically. He says (loc. cit.): "A
work of art always has aesthetic significance . . . whether or not it
serves some practical purpose, and whether it is good or bad, it de-
mands to be experienced aesthetically." . . . "Where the sphere of
practical objects ends, and that of 'art' begins, depends, then, on the
'intention' of the creators." (op. cit., p. 12.)
Panofsky's remarks emphasize the necessity for looking for the
way in which aesthetic elements operate within the art object and
the way in which the art object operates in its physical context, in
contrast to and in addition to the ways in which the object works
mechanically and is used socially. In the case of visual art — sculp-
ture, painting, and, partly, architecture — the material form is so
structured that the object may be said to function aesthetically, by
focussing and channelling visual attention.
The novelist E. M. Forster (1949, p. 31) argues that the phrase
"art for art's sake" should be held to mean primarily that "a work
of art — whatever else it may be — is a self-contained entity, with a
life of its own imposed on it by its creator." He goes on to say
I, 6 ±~
LEWIS: DEFINITION OF PRIMITIVE ART 223
(loc. cit.) : "It has internal order. It may have external form. That
is how we recognize it." Forster discusses order in various spheres
of life and says that perfect order or perfection in daily life, in sci-
ence, in history, and in political life, is unattainable. However,
works of art "are the only objects in the material universe to pos-
sess internal order." (op. cit., p. 34.)
This characteristic of art objects — the total design based upon
order, consistency, rhythm, harmony and balance of the lines, shapes,
masses, colors, light values and textures — enables us to recognize art
quality in an item of material culture. When perceived visually,
the form of art objects is more ordered, consistent, rhythmic, har-
monic and balanced than that of non-art objects.
Art objects, suggests Forster, can be, and often are, "perfect."
The absence of perfection in other spheres of life, as in science or in
politics, stems from the nature of the elements with which scientists
and political leaders have to work. Scientists work in an open-ended
universe, of which only fragments are amenable to control. In polit-
ical life, the notorious intractability of humans destroys elegantly
conceived political schemes. In art, however, the artist has much
more control over the elements of his creation. He can and does
arbitrarily set limits to the projected work, chooses his tools and
materials, and not only is able to operate within the conditions of
work but is often able to create those very conditions. In the case
of visual art, and especially in the more ephemeral arts of dance and
- music, the art object is intended to produce a specific effect, often
sharply limited in time and space. A scientific or political system
is of little importance unless it can persist and unless it will be valid
under varying conditions.
The characteristic of an artifact which makes it recognizable as
art is the aesthetic organization of its visible form. Both Panofsky
and Forster emphasize the intention of the artist to make artifacts
that will be aesthetically important. The artist thus works purpose-
fully and consciously to achieve aesthetic organization of a work of
art. However, let us now consider why art is made, by asking, what
is it that art does?
Art provides one way of presenting symbols for use in the social
life of man, and of fixing in visible and persisting form otherwise
intangible and transient images, ideas and events. In visual arts,
persistence of art objects, barring purposeful destruction of the ob-
jects, is limited only by the durability of the materials out of which
the object is made. In music and dance, the art object, for instance,
FIELDIANA: ANTHROPOLOGY, VOLUME 36
Fig. 108. Hanger made of antlers of a deer, Cervus (Rusa) philippinus; early
20th century. Apayao, Luzon, Philippine Islands. Height 11 Yi inches. CNHM
Cat. No. 109499.
a performance, may persist only as long as the performance lasts.
But the art quality — the harmony, repetition, contrast, and rhythm
— both of durable and of transient art, helps fix the symbols in the
minds and memories of onlookers, at the moment of presentation,
by contrasting these symbols with the everyday round of life and
work. Art as an important part of ceremonial is a conscious and
determined effort to make important symbols visible and permanent.
LEWIS: DEFINITION OF PRIMITIVE ART
Fig. 109. Hanger, carved wood; early 20th century. Apayao, Luzon, Philip-
pine Islands. Height 13 inches. CNHM Cat. No. 109428.
It is possible to distinguish art from non-art on two levels. First,
by looking to form alone and with no specific knowledge of the life
of the people, we may seek evidence of design and composition for
visual appeal. At a second and deeper level, we seek to know the
artist's intent, function and meaning, using knowledge of the culture
and the society, in order to see how much of the total function of the
object is involved with visual appeal. Reference to form alone, as
in undocumented museum specimens, enables us to make a judgment
FIELDIANA: ANTHROPOLOGY, VOLUME 36
as to whether there is any art quality present; reference to intent,
function and meaning enables us to assess the amount of that qual-
ity present in a given art object. Art quality can be best measured
Fig. 110. Effigy hanger, carved and painted wood; early 20th century. Berlin
Harbor, New Guinea. Height 18 H inches. CNHM Cat. No. 140040.
when we know the artist's intent, and the meaning and function of
a work of art.
Figures 108-110 show three artifacts, of similar mechanical func-
tion, which served as hangers from which articles could be suspended.
Figure 108 illustrates a set of deer antlers used for this purpose. Fig-
ure 109 shows a hanger that was carved from wood. The presence
of regular notches on the arms could be construed as decoration, but
it could be argued that the notches provided better gripping surfaces
for tying objects to the hanger. In the third object, pictured in fig-
ure 110, there can be no doubt that art quality is present. The
Fig. 111. A, Adze, shell blade set in wooden handle, rattan binding; early
20th century. Cape Gordon, New Guinea. Height 28 \i inches. CNHM Cat. No.
B, Ceremonial adze, stone blade, carved wooden handle, sennet binding; early
20th century. Mangaia, Cook Islands, Polynesia. Total height 27 inches.
CNHM Cat. No. 111353.
228 FIELDIANA: ANTHROPOLOGY, VOLUME 36
physical form was primarily determined by the maker's intention
that a visual image be presented, in combination with the mechan-
ically functional hooks. Two adzes are shown, one in figure 111, A,
an undecorated and unelaborated tool, the other in figure 111, B, a
visually elaborated symbol.
All these objects were made either as implements or tools, or as
decorated or elaborated versions of such devices. The figure illus-
trated in figure 112 has no mechanical function whatsoever, but is
just as "useful" as the others. It was designed and carved to be set
up near an Ovimbundu blacksmith's forge to help insure successful
prosecution of that craft.
Some of the differences between art and non-art objects are thus
readily discernible.*' Non-art forms (figs. 108, 109, 111, A) are either
natural forms not altered by man, or are devoid of decorative or
otherwise visually attractive elements. It should be noted that
neither development and evolution from non-art to art nor the re-
verse is implied. All of the objects used to illustrate these differ-
ences are contemporary in time, and although drawn from different
societies, some could exist side by side in the same group.
SOME IMPLICATIONS OF THE TERM "PRIMITIVE"
• The dictionary definition of "primitive" stresses the original, first
or root stage of a development. The problem of dealing with con-
temporary non-civilized peoples must be solved, at least provision-
ally. Ralph Linton (1958, p. 9) says:
Actually, there is no culture extant today which can be regarded as primitive.
There are cultures of greater or less complexity, and cultures which have a greater
or smaller number of features in common with our own, but none of them are an-
cestral to the high cultures which we call civilizations. The way of life of an
American Indian tribe or a group of Polynesian Islanders does not represent a
stage through which our own ancestors passed any more than a modern dog repre-
sents a stage in the evolution of the elephant. Every existing culture has had its
own more or less independent evolutionary history.
Tax (1960, p. 441) objects to the use of the word "primitive,"
saying that according to the "dictionary meanings of 'primitive,'
there are no such groups — or, at least, the peoples that we study are
not primitive as [anthropologists] understand the term." Mednick
(1960) gives the dictionary definition of "primitive" and notes the
disparity between it and anthropological usage of the term.
* Linton, Tax and Mednick, basically, objected to the application
of the term "primitive," with its meaning of "early," "original," or
Fig. 112. Figure, carved wood, painted red and black. In former days a
blacksmith had to kill a victim, whose spirit, it was thought, entered a figure of
this kind, which was then made to stand by the forge; early 20th century. Ovim-
bundu, Elende, Angola, Africa. Height 14 J^ inches. CNHM Cat. No. 208338.
230 FIELDIANA: ANTHROPOLOGY, VOLUME 36
"first," to contemporary groups which, very obviously, are not early
or original. Linton objected, also, to viewing primitive contem-
poraries as representing an arrested stage of development through
which our civilization passed. However, if, as Linton said above,
". . . every existing culture has had its own more or less independent
evolutionary history," it follows that every society, present-day or
ancient, whether civilized or "primitive," at the start of its history
had a beginning stage which can be called, without equivocation,
Redfield's characterization (1953, pp. 6-14, 22) of primitive so-
ciety as a base out of which civilization may develop, presents a social
typology — an ideal primitive society and an ideal civilization. It
would be possible to view a given civilized society as having devel-
oped from its primitive beginning to its condition of civilization.
Also, one could measure degrees of primitiveness, most at the begin-
ning, less with more development, until at some point the society
could be called civilized. We could refer to all degrees of primitive-
ness as primitive, and to all degrees of civilization as civilized. We
never have the first stage of primitiveness available for study and
probably we never shall. Therefore, to reserve the term "primitive"
for application only to such beginnings seems wrong. If it is neces-
sary to have such a term, "primeval" could be used to refer to the
actual beginning. Mednick's statement (1960, p. 444) that Tylor
and the British Evolutionists used "primitive" to refer to "that long-
gone original population of mankind about whom one could only
speculate" should remind us that we can use the word "primitive"
as though it means "pertaining to" the original stage, as well as using
it to mean "original." Indeed, a dictionary definition of the word
gives both meanings: "1. of, or pertaining to the beginning," etc.,
and "3. original, first," etc.
For example, to refer to the art of Melanesia as primitive art,
although our knowledge of Melanesian societies is almost com-
pletely ethnological and extends back only about one hundred years,
implies that we have speculated about the nature of the antecedents
of contemporary Melanesian societies and have judged the recent
societies to be more similar to the ideal type of primitive society
than to an ideal type of civilized society. By and large, Melanesian
societies of today can be described in terms of Redfield's character-
istics of primitive societies rather than those of civilized ones, in spite
of the fact that they are much changed as a result of contact with
Europeans, and have had long (but unknown) histories. Also, in
LEWIS: DEFINITION OF PRIMITIVE ART 231
the absence of archaeological knowledge of Melanesia, we can only
speculate about the culture history of specific Melanesian societies.
Thus, on the basis of present knowledge, it is not unrealistic or
impractical to view Melanesian societies as primitive, in the sense
that they have not been and still are not civilized, and also in the
sense that as far as anyone knows they have not changed much
from the actual primitive beginning. Future archaeology could show
that the first Melanesian societies, of thousands of years ago, were
smaller, more isolated, etc., than the present ones, and we could
then measure the change; but it is theoretically possible (although
improbable) that earlier Melanesian societies were less primitive in
social organization, which could be the case if they were formed by
Indonesian or mainland Southeast Asian migrants overrunning and
combining with Australoid groups in the area.
Art historians sometimes refer to paintings of the twelfth, thir-
teenth, and early fourteenth centuries in Italy as primitive Italian
paintings; that is, this period is considered an early stage of the
Renaissance. Other painting styles existed in Italy prior to that
period, for instance, Byzantine and Roman, but these are not con-
sidered to be part of the Renaissance.
Thus it can be seen in art historians' use of "primitive" that the
frame of reference is important. For the historian of the art of the
Renaissance, the twelfth century is a beginning stage, and therefore
primitive. Note, however, the inclusion of almost three centuries as
a beginning stage! For an anthropological investigator of art, study-
ing a worldwide range of society from all ages, the frame of reference
must be universally applicable.
Gerbrands (1957) devotes a whole chapter to an attempt to des-
ignate for primitive art a term which applies to "the art of the au-
tochthonous population of Negro-Africa, Australia, the Pacific area,
and certain forms of art in America, Asia and Indonesia." (op. cit.,
p. 9.) As a working term he proposes the name, "non-European
art." He (op. cit., pp. 11-12) has several objections to the use of
"primitive": (1) that chronological priority is implied; (2) that
"primitive" has connotations of crudeness or clumsiness; (3) that
recent and present-day societies labeled "primitive" are thought of
as arrested stages in the development of European culture. Also,
and confusingly, in history of art, (4) certain pre-Renaissance art
styles are called primitive, for example, Flemish or Italian primi-
tives; and (5) works of certain modern painters, for example, Henri
Rousseau, are called primitive. Because of these meanings and con-
'WIVERSfTY Of u.m
232 FIELDIANA: ANTHROPOLOGY, VOLUME 36
notations, and because of the several confusing ways in which "prim-
itive" is used, Gerbrands suggests the name non-European art. This
term leaves much to be desired. Gerbrands would exclude from this
field of study European Paleolithic art, a body of material which
could most appropriately be called primitive art. He says (op. cit.,
p. 11) : "Primitive can certainly be used to indicate the initial stage
of human culture as such, but then it must be understood as referring
to the culture of Paleolithic man, which does indeed, as far as we know
at present, represent the earliest phase in cultural development."
Another name suggested as an alternative for primitive art is
ethnological art. An argument for this name was given by Hasel-
berger (1961, pp. 341, 342). ' A major objection to the name is the
exclusion of archaeologically recovered art from consideration as
primitive art. Although archaeological research methods are dif-
ferent from those used in ethnology, and although different kinds of
materials and knowledge about them are collected in the two disci-
plines, it does not follow that the art or the society studied differs
in any significant way. Thus, to arbitrarily exclude an archaeolog-
ical dimension from the study of primitive art would mean that the
study would lack time depth. This would exclude possibilities for
study of change in art.
Gerbrands limits his study to the art of indigenous peoples of
Negro Africa, Australia, the Pacific Islands, and certain forms of art
in America, Asia and Indonesia. From Asia and Indonesia he ex-
cludes Buddhist and Hindu art of South, Southeast and East Asia,
Mohammedan art, and also, rather anticlimactically, Japanese prints.
The archaeological art of America he considers "a dubious case" (op.
cit., p. 137), as well he might, for as a category it contains the art of
civilized, pre-civilized and non-civilized societies.
Gerbrands' non-European art is a smaller category than that be-
ing proposed here as primitive art. Non-European art seems to in-
clude only the art of ethnologically known peoples; primitive art, as
defined herein, includes, in addition, the art of archaeologically known
peoples. Gerbrands' conclusion is that neither form nor content of
non-European art can be subsumed under a single principle of simi-
larity. He (op. cit., p. 138) cites Leonhard Adam (1954, p. 30), who
said about the disparate arts of the primitive societies that their
"foreignness in form and content serves to link them together in
our mind for the purposes of art criticism. The link ... is extra-
neous to the works themselves. It depends upon us and our atti-
tude to them."
LEWIS: DEFINITION OF PRIMITIVE ART 233
Differences in form and content stem from differences in culture.
The similarities within the category of primitive art lie in the simi-
larities of the social context in which the art is produced and used.
Some implications for determination of art form stemming from the
social context of art are discussed below. Also, technology of art,
including techniques and use of materials, has a determining effect
upon form. Finally, as an aspect of form, the size of art objects yields
yet another principle of comparison showing similarities within prim-
itive art which contrast with the art of civilized societies. Actual
comparison of aspects of form of the art of primitive and civilized
societies will be undertaken in another paper.
Although the term "primitive" has some derogatory connotations
and has to be explained and defended, it seems wrong to abandon
or change the word in favor of others which are less pertinent and
less accurate. The term "primitive," when used in history of art
and in art criticism, is usually complimentary. Primitive art is
eagerly sought today by museums and private collectors and is very
highly valued. There is no question of patronizing connotation in the
term. Rather, one might ask whether there has not been overly
complimentary evaluation, both aesthetic and financial, of many ob-
jects originating in primitive societies.
IMPLICATIONS FOR ART IN DIFFERENCES BETWEEN
PRIMITIVE AND CIVILIZED SOCIETIES
As has been proposed above, primitive art is art which is pro-
duced and used by members of primitive societies. Redfield (1953,
pp. 6-14) considered as primitive those societies whose communities
were small, isolated, homogeneous and intimate, where literacy was
absent, where full-time specialists were lacking, where there was in-
formal social control, and where there was strong group solidarity
based upon common understandings as to the purpose of life. On
the other hand, Redfield (op. cit., p. 22) held that civilization may
be said to exist to the extent and degree to which, and in the respects
in which a society has developed away from these characteristics of
primitive society. It is the kind of society characterized by Red-
field as primitive which has been taken here as an ideal model for
the type of society which produces primitive art.
Three of the characteristics of primitive society cited by Redfield
can be considered to be inter-related: smallness, homogeneity, and
intimacy. Mere size would not have much implication for art, but
the attendant homogeneity and intimacy surely do. Homogeneity
234 FIELDIANA: ANTHROPOLOGY, VOLUME 36
of language, of thought, and of the ways to conduct life make it
certain that the artist is the same kind of social being as other mem-
bers of his society. The relative lack of divisions in primitive soci-
eties insures this. * Primitive artists usually are fully integrated mem-
bers of homogeneous and intimate societies; our Western European
artists often are marginal men — Bohemians — and as such are consid-
erably divorced from the everyday life of other members of society.
/ A marked characteristic of the smallness, homogeneity and inti-
macy of primitive society is the lack of full-time specialists — spe-
cifically for art, the lack of full-time specialist-artists. One implica-
tion of the use of part-time artists is the relative lack of experience.
A full-time artist has maximum opportunity to develop his crafts-
manship — to become more deeply immersed in his art than a part-
time artist. This means that a full-time artist has the opportunity
to become more skilled than a part-time one; but also it may be that
the full-time specialist can slip into habitual and repetitive expression
in his art, while the part-time artist's work may be cruder, but fresh-
er and more vital.
There are implications for change or lack of change in art stem-
ming from specialization of artists. When full-time artists produce
art, there is more tendency for training schools to arise — for codifi-
cation of methods and techniques, which exert stabilizing influences
on changes in art styles and help insure uniformity of expression.
The expression of part-time artists, on the other hand, is controlled
by the ideas of the broader, often the whole, adult group — not by a
small and professional segment of society. Isolation of primitive
societies, however, makes it less likely that alien art influences are
felt, so that primitive art styles have considerable stability.
i Art techniques in which complex technological processes and com-
plicated social organization are necessary often require complex divi-
sion of artistic labor. For instance, to make a bronze casting in lost
wax technique it is necessary to have supplies of suitable metal, to
have access to wax for the model, to have access to molding mater-
ials, to have equipment and knowledgeable assistants for the foun-
dry operations. Large-sized sculpture, architectural sculpture and
paintings, and large buildings, especially those made of durable ma-
terials, are done by professional and specialized artists and architects,
who draw the plans and then often depend upon engineers, foremen
and overseers and a supply of skilled and unskilled labor to carry out
the execution of the designs. Organization of communication, sources
of supply, and financing are also complex compared to undertakings
LEWIS: DEFINITION OF PRIMITIVE ART 235
of similar nature (if comparable at all) in primitive societies. The
informal social control of primitive societies is not adequate for the
organization and sustained execution of vast enterprises such as
building cathedrals, pyramids and similar great works.
The absence of literacy in primitive societies has implications for
the production of art. Written canons of design and instructions
for artists are features of civilized societies. But more important
than lack of literate communication about art is the possibility that
literacy in civilized society takes over some of the functions which
art serves in primitive societies. In literate civilized societies, the
books, newspapers, and written inscriptions on monuments and sign
posts facilitate communication by fixing knowledge and information
about objects, events, and concepts in material form so that such
knowledge persists through time and can be transported across space.
In primitive societies art objects accomplish some of these same func-
tions by fixing knowledge and information in the material form of
statues, paintings, and decorated objects, which also can persist
through time and sometimes can travel in space. It is interesting
to note here that written language, especially that known to us from
monuments and other documents, often utilizes visual composition
and design elements in the lettering or typography, and thus may be
considered to be a specialized form of visual art.
But so far we have been discussing structural aspects of primitive
society, such as small size, lack of specialization, and homogeneity —
all part of what Redfield (1953, p. 21) called the technical order,
". . . that order which results from mutual usefulness, from delib-
erate coercion, or from the mere utilization of the same means."
Less obvious, less tangible, and less amenable to investigations, but
nevertheless very important as a set of determining forces for art is
what Redfield (op. cit., p. 20) called the moral order: "All the bind-
ing together of men through implicit convictions as to what is right,
through explicit ideals, or through similarities of conscience. The
moral order is therefore always based on what is peculiarly human —
sentiments, morality, conscience — and . . . arises in the groups where
people are intimately associated with one another."
Civilization, Redfield stated (op. cit., p. 22), may be thought of
as that society in which the relations between technical order and
moral order take forms radically different from the relationships be-
tween the two in pre-civilized society. Also, he said (op. cit., p. 23)
that in primitive societies the moral order is great and technical
order is small, and that civilization greatly develops the technical or-
236 FIELDIANA: ANTHROPOLOGY, VOLUME 36
der, including formal regulations of state and church and non-moral
ordering of market place behavior. Thus, as civilization develops,
the earlier primitive moral order is broken or altered. A new and
more inclusive moral order may then arise in the civilization.
Since in primitive society art is very much involved with religion,
therefore with a generally sacred sphere of life, we look to art as
being symbolic of many aspects of the sacred and moral order. Also,
just as there is sharing of sentiment and conscience as to what is
right in other aspects of life, so there is with respect to what is right
An artist at work in a primitive society is limited by the isola-
tion of his group, so that he is unaware of alternative styles, but
he may have positive affection and sentiments toward his people's
traditional style. In a situation of culture contact or other social
change, conflicting sentiments and searchings of conscience arise;
choice between alternatives then becomes necessary. As a result,
in changing societies art styles change, different factions in art arise,
some styles are distributed widely and copied, and others remain
local. Thus, although sentiments and conscience of members of a
given society are relatively difficult to deal with, their presence
should be noted as important for study of art, in that they are fac-
tors in determination of art form. Differences in the moral order
mean differences in the art.
A similarly intangible aspect of life is world view, which also is
important for art. Some of the elements universally present in the
world view of all peoples, Redfield said (op. cit., p. 92), are the "We-
They" difference, which would include arrangements of humans on
a universal stage, the "Man-not-Man" difference, including con-
ceptions of human nature, the spatial and temporal organization of
the universe, and the universal human experiences. All such ele-
ments of world view are important as subject matter for art, and
thus are, in part, factors determining art form; they also provide
indices with which to compare societies and art forms. Differences
in world view mean differences in art.
It has been suggested here that the term "primitive art" can be
used to refer to the art of certain societies which are considered to
be typologically primitive. The main meaning of the dictionary defi-
nition (primitive= original or early) is not incompatible with its use
by art historians to refer to an early stage of a development and is
quite in accord with Redfield's primitive-civilized dichotomy. There
do seem to be differences, in kind and in degree, between societies
that are primitive and those that are civilized.
LEWIS: DEFINITION OF PRIMITIVE ART 237
Since art forms, and therefore art styles, are largely determined
by the nature of their socio-cultural contexts, it follows that the art
of primitive and civilized societies should reflect differences in such
Redfield's scheme was developmental in that he thought that
civilization develops out of, and away from, primitive society, and
this suggests that the art systems comprising a total body of primi-
tive art can have an historical dimension; thus, the world's primitive
art can be defined in space and time. Let us now proceed to do just
that, by surveying the world and indicating which societies produce
THE PRIMITIVE SOCIETIES
One objection raised by Tax and Mednick against the use of the
term "primitive" is that societies of considerable complexity, i.e.,
civilized societies, are often included in various groupings labeled
"primitive." To meet this objection it is proposed that societies
that are typologically civilized should be excluded from consideration
as primitive societies. With respect to most of the civilized societies
of the world there is no difficulty in recognizing them as such and
excluding them. However, there are several areas of the world which
present special problems: Africa, Nuclear America and Eurasia.
What is the relationship between civilized and primitive societies
in these areas?
The rise of civilization in Egypt is considered to be part of the
rise of civilization in the Mediterranean basin. In Africa south of
the Sahara — one of the most important source areas for primitive
art — there were developments of civilization, particularly in West
Africa, which demand that certain West African societies be excluded
as sources of primitive art. J. D. Fage (1959) writes of "the great
empires of the western Sudan," and of "the great states of the Guinea
forest." He tells of the existence of various western Sudanese "em-
pires," some Islamic, some pagan, such as ancient Ghana and Mali,
Songhai and others, and also the states of the Guinea forest — Oyo,
Benin, Dahomey and Ashanti. Murdock (1959, pp. 64-68) postu-
lates for Negro Africa an indigenous invention of agriculture which
occurred around the headwaters of the Niger in the western Sudan,
and says: "It is probably no accident that the earliest and most com-
plex civilizations in this part of Africa of which we possess actual
historical records, . . . were exhibited by Mande-speaking peoples."
The civilizations to which Murdock refers are the western Sudanese
"empires" listed by Fage.
238 FIELDIANA: ANTHROPOLOGY, VOLUME 36
Thus, in West Africa there were groups which displayed a number
of the characteristics of civilized society, such as large populations,
social stratification (including kingship and slavery), formalized po-
litical, military and religious organization, urbanization, and literacy.
The relationship of present-day ethnologically known Sudanic
tribes to the historically known Sudanese states is complex and not
well known. All these tribes are to some degree changed from the
conditions of primitive society by their contacts with such states, from
interdependence due to trade, conquest, or displacement from habitat
areas. The societies referred to by Fage as forest states — Ashanti,
Benin, Oyo (Yoruba) and Dahomey — display similar complexities
of relationship but are somewhat simpler to deal with because their
periods of political ascendancy were more recent. The art of Ashanti,
Dahomey, Oyo and Benin is always included in studies of African
art and is often considered to be primitive art. It is extremely in-
teresting to note that among these last-named civilized societies are
found art techniques and forms that sharply contrast with other and
primitive African art-producing societies. Cire perdue metal casting
is used in all four groups and is rarely found elsewhere in Africa ex-
cept in closely adjacent areas, as, for instance, among the Baule, who
make brass weights for weighing gold, and among some of the tribes
of the Cameroons grasslands, who do brass casting. The naturalism
of the Ife bronzes and terra cottas, and of the (presumably related)
Benin bronzes contrasts greatly with other African art. (This note
on technique and form is not presented as evidence of civilization;
rather, it anticipates comparison of art forms of primitive and civi-
- On the basis of the presence of complex social organization, these
several Guinea forest states and western Sudan states are classified
as civilized societies, and as such are excluded from the classification
of primitive. Their art is not primitive art.
In the Americas, the art of civilized societies is often included in
exhibitions and collections labeled primitive art. My colleague,
Donald Collier, discussed the relationship between primitive and
civilized American Indian societies in the catalogue for the ex-
hibition of Indian Art of the Americas held in 1959 at Chicago
Natural History Museum. He considered the arts of the Maya,
Aztec, Inea and other of the great Indian societies as being not prim-
itive, but as "best understood and appreciated when considered on a
level with the arts of the ancient Mediterranean and Oriental civili-
zations." Also, he referrred to the Circum-Caribbean tribes of Cen-
LEWIS: DEFINITION OF PRIMITIVE ART 239
tral America, the Antilles and parts of Colombia and Venezuela as
... an intermediate position developmental^ and geographically between the
primitive and civilized Indian societies. . . . These farming Indians were grouped
in social classes and ruled by powerful chiefs. . . . Their culture and art were
strongly influenced by their civilized neighbors in Middle America and particu-
larly by those in the northern and central Andes. . . . Unlike the primitive tribes,
they produced a great variety of gold and copper ornaments by casting and other
complex techniques. These metallurgical techniques as well as stylistic traits were
diffused from the Andean civilizations to the Circum-Caribbean tribes in Central
America during the first millennium after Christ ....
Although it is useful to distinguish between the arts of the primitive and
civilized societies in the New World, it is also important to remember that these
societies were not completely isolated from one another and that they were intri-
cately connected by historical relationships of great time-depth. As in other
areas of the world, the art styles of the nuclear areas of civilization exerted power-
ful influences on the arts of the marginal and less developed societies. . (Col-
Thus in both Africa and in the Americas the art of certain societies
usually included in the realm of primitive art should be excluded on
the grounds that these societies are typologically civilized. There
would be little or no argument about the civilized nature of the so-
ciety of some of these groups, as, for instance, Ashanti, Dahomey,
Oyo, Benin, Ghana, Mali and Songhai in Africa, and Inca, Maya,
and Aztec in the Americas. However, difficulties in classification
arise with many other societies related to the civilized groups by
association, conquest, or ancestry. In other words, we can easily
classify some societies as civilized, and thus remove from the cate-
gory of primitive art some styles and techniques that have seemed
out of place; but with a number of societies related to, or neighbors
of, centers of civilization, much more investigation is necessary to
definitively establish their relationship to civilization.
With respect to Europe and Asia, I can but point to some broad
and obvious matters. First, we may logically assume that primitive
societies existed prior to the development of civilization. Secondly,
compared to other areas of the world, there have been relatively few
survivals of primitive societies in Eurasia — another way of saying
that European and Asiatic civilizations have been flourishing longer
and have spread more widely than civilizations in other parts of the
world. This means that there are relatively few examples of primi-
tive art from these areas. The Paleolithic cave and portable art of
Europe is primitive art, and some Neolithic art may be, also. The
240 FIELDIANA: ANTHROPOLOGY, VOLUME 36
classification of the latter depends upon knowledge of the develop-
ment of civilization in specific areas.
If these various past and present civilized societies of the world
are thus excluded from consideration as originators of primitive art,
which groups are left? In Africa they are the various indigenous
peoples of the past and the present, including Pygmy and Bushman,
and those Negro groups not involved in the developments of civiliza-
tion mentioned above. In the Americas they are the various Amer-
ican Indian societies of North and South America not involved with
the above-mentioned developments of civilization. In Oceania the
primitive societies are those indigenous to Australia, Melanesia,
Micronesia, Polynesia and Malaysia, in which latter area should be
excluded the extensions of Asiatic civilization into Indonesia. In
mainland Asia, exclusive of the Chinese and Indian civilizations,
which have exerted tremendous influence, several groups may be
considered as primitive societies: the various Siberian tribes, the
Ainu, various central Asian nomads and some of the marginal abo-
riginal peoples of south China and mainland southeast Asia.
It has been suggested herein that art can be differentiated from
non-art by the presence of elements of form whose total design em-
phasizes visual appeal, and by reference to the use to which the
maker intended the object to be put. Art objects can be defined as
artifacts that function primarily by means of appeal to the visual
sense. In decorative art, the function of the object would be partly
mechanical, partly visual.
Primitive art is defined as the art of societies that can be regarded
as primitive by virtue of type of social organization. The term
"primitive," although it carries certain invidious connotations and
has some confusing aspects, is appropriate, if used to refer to an
ideal type of early society and then extended to later societies of
Art form has not been used in this paper as a criterion of primi-
tiveness or civilization. Rather, the definition of art and the classi-
fication of societies as primitive and civilized are a prerequisite order-
ing of social contexts of art so that future comparative analysis of
art form can proceed.
1954. Primitive Art. Penguin Books.
1959. The Diversity of Indian Art. In Indian Art of the Americas. Chicago
Natural History Museum.
Face, J. D.
1959. An Introduction to the History of West Africa. Cambridge University
Forster, E. M.
1949. Art for Art's Sake. Harper's Magazine, August, pp. 31-34.
1936. Art through the Ages. (Revised ed.) Harcourt, Brace & Company.
Gerbrands, A. A.
1957. Art as an Element of Culture Especially in Negro-Africa. No. 12.
Mededelingen van het Rijksmuseum voor Volkenkunde, Leiden.
1961. Method of Studying Ethnological Art. Current Anthropology, vol. 2,
no. 4, pp. 341-384.
1958. "Primitive" Art. In The Sculpture of Africa, pp. 9-16. F. A. Praeger.
1960. Memorandum on the Use of Primitive. Current Anthropology, vol. 1,
Murdock, George P.
1959. Africa, Its Peoples and Their Culture History. McGraw-Hill Book
1955. Introduction: The History of Art as a Humanistic Discipline. In Mean-
ing in the Visual Arts, pp. 1-25. Doubleday Anchor Books.
1953. The Primitive World and its Transformations. Cornell University Press.
1960. "Primitive" Peoples. Current Anthropology, vol. 1, p. 441.